Корреляция физико-химических свойств с антиоксидантной активностью в фенолах и тиофенолах.

Introduction

Free radicals generated within the body fulfil several functions, such as participating in cell signaling or fighting pathogens. To balance and maintain the levels of free radicals, the body utilizes endogenous and exogenous antioxidants to scavenge the excess radicals when necessary. Elevated levels of free radicals and other reactive oxygen species (ROS) can trigger oxidative stress, leading to damage in various biomolecules, such as lipids, proteins and DNA, and potentially resulting in cellular death 1 – 4 . The challenge in combatting oxidative stress lies in the absence of obvious symptom development.

Oxidative stress has been found to be a common factor in a variety of diseases, such as Alzheimer’s disease, Parkinson’s disease, cardiovascular diseases or preeclampsia among many others 5 – 8 . The accumulation of cellular and DNA damage, along with disruptions to cell signaling from ROS, all lead to the development of these disorders 4 . To treat or manage these diseases natural and synthetic antioxidants are commonly applied in various therapies 9 .

Many exogenous antioxidants are isolated from plants, with polyphenols, coumarins, or flavonoids being some well-known examples 10 . The most common scaffold found in many naturally and synthetically derived antioxidants is that of phenols. In plant-based antioxidants, phenols are present in form of polyphenols, curcumin, and lignans 10 . The phenol moiety acts by single electron transfer (SET) or hydrogen atom transfer (HAT) mechanism to scavenge radicals via donating an electron or an H atom, and the resulting phenolic radical is resonance-stabilized due to the extended conjugated system(s) 11 , 12 . Phenols and polyphenols are extensively researched within the antioxidant community 13 – 15 .

Another important functional group with high antioxidant activity is thiols, the most common representative of which is glutathione, one of the main antioxidants used in the body to protect cells from ROS damage 16 , 17 . The cysteine –SH group is used to aid the reduction of hydrogen peroxide/organic peroxides by glutathione peroxidase, and glutathione levels are replenished by glutathione reductase using NADPH released from the pentose phosphate pathway 18 . Another –SH group containing compound class is that of thiophenols which is a safer alternative to phenols, and they scavenge radicals through the same mechanism. Due to the larger size of sulfur and increased acidity of thiophenols, they are more likely to donate their proton or hydrogen atom, making them potentially more effective in stabilizing radicals this way compared to their phenol counterparts 19 . Despite their lower toxicity and significant potential as radical scavengers, thiophenols are underutilized as antioxidants; only sporadic studies are available in the literature 20 – 22 .

The investigation of antioxidants through computational studies has proven pivotal in unraveling the mechanisms that underscore their effectiveness 23 , 24 , 25 . Simulations offer invaluable insights into both the thermodynamics and kinetics governing antioxidant reactions. An examination of electronic and thermodynamic parameters emerges as a potent approach for comprehending the stability and reactivity exhibited by these molecules.

This work aims to compare the radical scavenging abilities of phenols and their respective thiophenol counterparts by theoretical as well as by experimental means. The goal is to determine whether the radical scavenging abilities of the selected compounds correlate with the calculated physicochemical property values, thereby aiding in identifying crucial structural features that could be used to enhance the scavenging capabilities of the main scaffold by the appropriate selection of substituents. The chosen compounds possess a wide variety of substituents at different positions on the aromatic ring, generating diverse levels of radical scavenging. Due to their simple structures, these sets of compounds provide a facile opportunity for comparing their physical properties and radical scavenging abilities. This comparison will provide a starting point of a rational model for predicting the radical scavenging capacity of a compound based on its calculated properties and will ultimately guide the design of more effective synthetic antioxidants. As mentioned above, the SH-containing gluthathione is an endogenous antioxidant that fulfills a crucial role in maintaining the body’s redox homeostasis. Despite, other SH-bearing compounds are rarely studied for their similar potential. Our work is the first systematic investigation of these compounds and by considering their limited toxicity compared to phenols, thiophenols could present a great opportunity to design low toxicity highly active antioxidants that may contribute to the development of effective antioxidant therapies.

Materials and methods

Substituted phenol and thiophenol derivatives, organic solvents (ethanol and DMSO), salts for buffer solutions (Na 2 HPO 4 , NaH 2 PO 4 , and NaCl), potassium persulfate for 2,2’-azino-bis (3-ethylbenzothiazoline-6-sulphonic acid) (ABTS) radical generation, and radical compounds (DPPH, ABTS, and AAPH) were purchased from Sigma-Aldrich, Oakwood Chemicals or Thermo Fischer Scientific, and were used in experiments without further purification.

Radical scavenging assays

Widely accepted radical scavenging assays, including 2,2′-azino-bis(3-ethylbenzthiazoline-6-sulphonic acid) (ABTS) and 2,2-diphenyl-1-picrylhydrazyl (DPPH) 10 , were utilized to evaluate the scavenging capacity of the studied phenol and thiophenol compounds. Detailed experimental procedures are summarized in the first subchapter of the supplementary information (SI).

Computational details

Thermochemical properties, such as bond dissociation enthalpies, ionization potentials, and proton affinities of seventeen phenol derivatives and their thiophenol counterparts were determined based on well-established literature protocols using the Gaussian 09 program suite 26 . All density function (DFT) calculations were performed using the Becke three-parameter exchange with Lee, Yang, and Parr correlation (B3LYP) functional along with 6–311 + + G(d,p) basis set for gas phase calculations, and B3LYP/6–311 + + G(d,p) scrf = (smd, solvent = water) level of theory for the solvated ones 27 , 28 . This method has been chosen due to its robustness for such calculations especially for trend analysis, which has been carried out in this work. It has also been used in many similar other earlier works and the comparability of this work to those investigations was of primary objective 23 , 29 – 31 . In addition, no conclusions were drawn from individual energy data. After conducting complete molecular geometry optimization at the same theoretical level, vibrational frequencies were calculated to confirm that none of the optimized structures exhibited any imaginary frequencies. This ensures all structures correspond to real local minima on the potential energy surface. The radical or cationic electronic structures were calculated using the Kohn–Sham formalism with unrestricted spin DFT/UB3LYP. Subsequently, optimized structures were employed to compute molecular descriptors, like Fukui indices. The Fukui function and dual descriptor values were collected by the Multiwfn 32 software suite, at the B3LYP/6–311 + + G(d,p) level of theory. Further parameters were computed to ascertain potential experimental correlations between the descriptors of phenols, thiophenols, and their experimental radical scavenging activities, encompassing adiabatic bond dissociation enthalpies (BDE), ionization potential (IP), proton affinity (PA), electron affinity (EA) thermochemistry descriptors. Additionally, S–H and O–H bond distances, dipole moments, logP values, HOMO and LUMO orbital energies and HOMO–LUMO gap energies were calculated. HOMOs, LUMOs, their energies, and HOMO–LUMO gap values were calculated using both the B3LYP/6–311 + + G(d,p) and M06-2X/6–311 + + G(d,p) level of theories 33 .

Results and discussion

Seventeen phenol and benzenethiol analogues have been selected (Fig. 1 ) and their experimental antioxidant activity and theoretical physical–chemical properties were determined. The goal was to select compounds with substituents that cover a broad range of electronic and steric properties to assemble a representative sample set. These compounds, with their simple yet structurally versatile nature, were chosen to elucidate the impact of substitution patterns on the antioxidant effectiveness of these core structures.

The experimental evaluation of the antioxidant activity involved two widely used radical scavenging assays: the 2,2-diphenyl-1-picrylhydrazyl (DPPH) and the 2,2’-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) assays 34 , 35 . The analysis revealed a broad spectrum of biological activity among both benzenethiol and phenol derivatives, as depicted in Fig. 2 .

Overall, the phenol derivatives demonstrated higher scavenging activity. Specifically, compound 1 ( o -NH 2 ), 4 ( p -OMe), 5 ( o -NH 2 , p -Cl), 6 ( o,p -Me), 13 ( m,p -OMe) and 17 (naphthyl) exhibited notable performance in both DPPH and ABTS assays, surpassing the activity of reference antioxidant, Trolox to a significant extent. On the other hand, alternating scavenging activity was observed for analogue pairs 2–19 , 8–25 , 9–26 and 14–31 where the phenols showed better activity compared to the benzenethiol analogues in the ABTS but not in DPPH assay. The thiophenol derivatives 25 , 26 , and 31 displayed consistent results across both test environments. Among the benzenethiol derivatives, compounds 32 and 33 stood out, aligning well with the predictions results discussed later. To illustrate the differences in activity trends within ABTS and DPPH environments, Fig. 3 shows the comparison of the pairs of analogues. The activity difference for each pair is calculated by subtracting the mean Trolox equivalent antioxidant capacity of benzenethiol from that of phenol for each compound. For example, the y-value for compound pair # 1 is obtained by subtracting the mean Trolox equivalent of 2-aminothiophenol ( 18 ) from that of 2-aminophenol ( 1 ). Thus, a positive value indicates that the phenol is a more active antioxidant, a negative value means the same for the thiophenol analog. The comparison of the two graphs reveals distinct patterns: in the DPPH environment, phenol and thiophenol analogues exhibit roughly equal levels of good scavenging activity with about a 50–50 ratio. However, in the ABTS environment, phenols demonstrate significantly greater activity than thiophenols.

Although chain breaker or scavenger antioxidants 36 exhibit considerable structural diversity, they share similar reactivity patterns and possible mechanistic pathways. Two distinct mechanisms participate in antioxidant processes: a direct hydrogen atom transfer and a single electron transfer 37 . Depending on the sequence of these two events, we can categorize three common but nonexclusive radical scavenging mechanisms, hydrogen atom transfer (HAT), sequential electron transfer-proton transfer (SET-PT) and sequential proton loss-electron transfer (SPLET) 38 . It must also be noted that in addition to these most commonly discussed mechanisms several other specific mechanistic proposals have been described 39 – 41 . The commonality in antioxidant reactivity enables the application of consistent theoretical approaches to study any radical scavenger. Over the years, certain indices have been introduced to describe antioxidant activity quantitatively. After conducting experimental assays to determine the radical scavenging activity of the compounds, extensive DFT calculations were performed to ascertain their characteristics in the gas phase as well as in aqueous medium. The data are tabulated in Table 1 . Once the calculated values were obtained, they were plotted against the experimental activities to explore any meaningful correlations between the physical characteristics and the activity. The plots showing significant relationships are displayed in Fig. 4 , while all results can be found in Fig. S1 in the SI.

The correlations depicted in Fig. 4 are selected from the broad array of calculated data on the basis of providing the best fit to the descriptive functions. The experimental data were plotted against all calculated parameters, considering linear, inverse, quadratic, cubic and quartic relationships and although Fig. 4 only represents the best matches (as described by their R 2 values), all correlations have been shown in Tables S3 and S4 .

One essential molecular descriptor is the bond dissociation enthalpies (BDE), which represents the energy required for homolytic cleavage of O–H or S–H bonds. A lower BDE indicates increased activity of the corresponding –OH or –SH group in the HAT mechanism, resulting in the formation of a more stable radical. In most cases, benzenethiols exhibited lower bond dissociation enthalpies (BDE) values in gas phase calculations, given sulfur’s larger size relative to oxygen that provides additional stability for the forming radical. One must, however, note, that mechanisms proposing H-abstraction include hydrogen atom transfer (HAT) and proton-coupled electron transfer (PCET) 42 as well. Since in our investigations, no unambiguous evidence was found that the radical scavenging activity followed the HAT process; hence, we will refer to H-abstraction as ‘formal HAT’. Nonetheless, there were exceptions to this pattern. The BDE values for derivatives 6 , 8 , 9 , 14 , 16 , and 17 indicate a more facile H atom transfer from the phenol than from its thiophenol counterpart (Table 1 , Fig. 5 ). Despite these observations, a strong correlation between the BDE values and the experimental antioxidant activity was only observed for thiophenols in ABTS assays (Fig. 4 h, Table S4 ).

Considering electronic and thermodynamic parameters is an effective way to understand the stability and reactivity of molecules. The reactivity is significantly influenced by the medium’s polarity 43 . For instance, hydrogen atom transfer typically results in neutral species and is therefore more favorable in nonpolar solvents. In contrast, SPLET and SET-PT mechanisms are generally preferred in polar environments due to the inadequate solvation provided by nonpolar environments for ionic species, which hinders the favorable progression of the radical scavenging action 44 . The literature frequently addresses the calculation of physicochemical properties in both gas phase and aqueous medium. Accordingly, this study also investigates the impact of solvation. Regarding the BDE values, they did not show significant change in trend as a result of applying the solvation model (Fig. 5 ), unlike the proton affinity and ionization potential values (Table 1 ).

The stability of an anion can be inferred from the energy necessary to break a bond via heterolytic fission. When the proton affinity (PA) value is lower, it implies a greater tendency for the -OH or -SH group to undergo deprotonation. Proton affinity is determined as the negative of the enthalpy change associated with the gas-phase reaction of a neutral chemical species with a proton, resulting in the formation of the conjugate acid of the former. The collected PA results demonstrated that the benzenethiolate anion is theoretically more stable than its phenolate counterparts, except for compound 16 (Fig. 6 and Table 1 ). Proton affinity is one of the main intrinsic reactivity indices of mechanism SPLET. In the water-solvated model, the PA values are notably lower than both the BDEs and IPs listed in Table 1 , suggesting that SPLET is likely the predominant mechanism in aqueous environment. However, a significant correlation between the PA values and experimental scavenging activity was not found (Tables S3 – S6 ).

Ionization potential (IP), or ionization energy, describes how easily a given compound contributes an electron, or in other words, how easily it can be oxidized. This property is often connected to the SET-PT mechanism. A lower IP of a substance suggests a higher likelihood of providing antioxidant protection through electron donation, meaning there is a greater chance for the substance to act as an antioxidant and donate electrons to guard against oxidative damage. Apart from compounds 1 and 5 , benzenethiols demonstrated a greater tendency for electron donation in the gas phase (Table 1 ). This trend significantly shifted in favor of phenols in the aqueous model (Fig. 7 ). This means that nine additional phenols are predicted to transfer electrons more easily in aqueous medium than their corresponding benzenethiols. The IP values of phenols exhibited a strong correlation with experimental results in both gas phase and aqueous medium ( R 2 = 0.82 (gas phase) and R 2 = 0.73 (aqueous model)), as illustrated in Fig. 4 a, e. Conversely, the trend for thiophenols revealed a more complex correlation pattern, showing moderate correlation in the gas phase (R 2 = 0.5, Fig. 4 b).

Finally, electron affinity (EA) measures a compound’s ability to attract and bind an electron. Good electron affinity of an antioxidant signifies its enhanced efficiency in electron capture compared to other compounds. This involves transferring an electron from a free radical to the LUMO of the antioxidant when it is in its neutral state. Regarding these values, we observed the largest discrepancy between calculated values in gas phase and water; thiophenols demonstrate higher theoretical efficiency of electron capture in gas phase, whereas in the aqueous model, the trend reverses, with phenols exhibiting greater electron susceptibility in most cases (Fig. 8 ).

Numerous studies within the literature have analyzed the substitution effects of simple para -substituted phenolic and thiophenolic compounds 20 – 22 , 45 . Our data are consistent with previously observed trends, where ionization potential in both gas and aqueous media indicated a decrease in electron donating affinity (higher IP) concurrent with an intensification of the electron withdrawing effect of the substituent (Fig. 9 and Table S2 ). Electron transport appears more favorable when an electron-donating group is attached to the benzene ring; this configuration helps stabilize the generated positive radical (OH •+ or SH •+ ). In contrast, the presence of electron-donating groups in aqueous medium increases the proton affinity (PA) of OH and SH bonds. We hypothesize that in these cases, the resulting anions (O - and S - ) are better stabilized by an electron-withdrawing group on the opposite side of the benzene ring, thereby enhancing proton transfer (see Fig. 9 ). Regarding the PA values collected in the gas phase, it is important to note that a larger dataset is necessary for more definitive conclusions. In comparison to reactions occurring in the gas phase, the presence of a solvent attenuates the impact of substituent effects on IP values 46 .

Within computational chemistry, the distribution, and the energies of frontier molecular orbitals (FMO) are imperative reactivity descriptors. HOMOs and LUMOs have significant roles in the chemical reactivity and active sites of a given compound. HOMO energy is proportional to the IP and describes the molecule’s affinity to accept an electrophilic attack. The lower the E HOMO of a compound, the worse its ability as an electron donor. Meanwhile, LUMO is linked to electron affinity (EA) and defines the molecule’s response to a nucleophilic attack. The HOMO–LUMO gap as parameter represents the ease with which an electron in a molecule can be excited from the HOMO to the LUMO. A lower value indicates that electron migration between these orbitals occurs more readily, leading to a faster radical reaction due to increased kinetic stability. Thus, multiple quantum chemical characteristics of the compounds were also calculated to identify further descriptors for the antioxidant activity. The data are tabulated in Table 2 , and the full list of HOMOs could be found in Table S2 in SI.

Among phenol derivatives those featuring electron-donating substituents—such as compounds 1 ( o -NH 2 ), 6 ( o, p -diMe), 8 ( o -Et), 13 ( m, p -diOMe), and 14 ( p-tBu )—demonstrate superior antioxidant activities in both assays, except compound 8 and 14 that only did well in the ABTS environment. Conversely, derivatives with electron-withdrawing substituents, including compound 7 ( m, p -diCl), 10 ( o -CF 3 ), and 12 ( o, p -diF), exhibit more moderate scavenging activity, and only within the ABTS assay (Fig. 2 ). These findings are further corroborated by predicted HOMO values (Table 2 ). A similar pattern is observed with thiophenols, where compounds 18 , 23 , 25 , 30 , and 31 , which possess electron-donating groups, have been found to be more effective scavengers compared to those with electron-withdrawing groups, such as compounds 24 , 27 , and 29 . Interestingly, pyridine-4-thiol ( 15 ) despite featuring an electron-withdrawing group inductively still exhibits one of the most promising activity profiles among thiophenols.

Our investigation primarily targets the HOMO orbital of antioxidant compounds, anticipating that prevalent free radicals, such as hydroxy ( • OH) and peroxy ( • OOR), act as electrophiles against the antioxidant moiety 36 , 50 , 51 . Figure 4 reveals a notable correlation between the HOMO energies and the scavenging activity profiles for both phenols ( R 2 = 0.62, Fig. 4 c) and thiophenols ( R 2 = 0.66, Fig. 4 d) in gas phase. Generally, thiophenol derivatives exhibit higher HOMO values, indicating a stronger antioxidant potential, except for compounds 1 , 4 , 5 , and 13 (Fig. 10 , left). These phenols also show robust scavenging activities (as depicted in Fig. 2 ), aligning well with the theoretical predictions. The trend undergoes a slight shift in calculations performed in aqueous medium, where compounds 12 , 14 , and 17 , alongside the phenol derivatives previously mentioned, are also predicted to possess elevated HOMO values (Fig. 10 , right). Similar to IP, strong (but reverse) correlations were observed in aqueous medium between the phenols’ HOMO values and their scavenging activity in the DPPH assay ( R 2 = 0.78, Fig. 4 f). However, no such correlation was found for thiophenols (Table S1 ).

Phenol and thiophenol derivatives that exhibit the highest scavenging activity are displayed and compared in Fig. 11 . This figure not only presents the theoretical HOMO energies but also illustrates the orientation of the molecular orbitals. A significant difference in the orientation of the orbitals is observed between analogues 13 – 30 and 15 – 32 . For compounds 13 and 30 , the HOMO primarily contributes to the delocalized π system in the benzene ring and the three oxygen atoms, although in compound 30 , the HOMO is more distributed around the thiol group and the connecting part of the benzene ring. In compound 32 , a higher HOMO contribution is also observed in the thiol group, compared to the hydroxyl group in compound 15 . Additionally, the distribution of the HOMO in compound 32 differs as it extends above and below the molecular plane, whereas in compound 15 , the HOMO aligns more closely with the plane of the molecule.

The indicated HOMO values follow the expected trend: compounds 1 and 13 have lower values than compounds 18 and 30 , respectively, and similarly, compounds 32 and 33 surpass their phenol counterparts. Please visit the supporting information for a comprehensive list of the HOMOs of the studied compounds.

In addition to mechanistic and virtual descriptors, the dipole moments, spin densities, bond lengths, and logP values have been determined (Table 2 ). The dipole moment is linked to the solubility, stability, and electron and hydrogen donation capabilities of the compounds. Generally, higher dipole moment increases solubility in polar solvents, such as water, potentially enhancing the effectiveness of antioxidants in polar biological environments. Moreover, a higher dipole moment may stabilize charge-separated states crucial in electron-transfer mechanisms. The highest dipole moments, measured in both the gas phase and aqueous medium, were observed for compounds 10 ( o -CF 3 ), 11 ( p -NO 2 ), and 16 (4-phenyl-2-hydroxyoxadiazole) among phenols, as well as their counterparts in thiophenols, compounds 27 , 28 , and 33 . The collected data indicates higher dipole moment values in aqueous medium for every phenol and nearly all thiophenol derivatives, except for compounds 21 ( p -OMe) and 30 ( m,p -diOMe). A significant correlation was found between the dipole moment and scavenging activity in phenols using the ABTS assay, both in the gas phase and aqueous medium, with R 2 values of 0.65 and 0.66, respectively (Fig. 4 i,j).

Regarding the O–H and S–H bond length values, similar trends were observed for phenols in both the gas phase and aqueous medium, with compound 16 (4-phenyl-2-hydroxyoxadiazole) exhibiting the longest bond length. In water, all phenols display longer bond lengths, as expected due to the presence of hydrogen bonds. Among thiophenols, the longest bonds were observed in compounds 18 ( o -NH 2 ), 22 ( o -NH 2 , p -Cl), and 33 (4-phenyl-oxadiazole-2-thiol). Greater discrepancies were noted when analyzing the trends for thiophenols within the derivatives in both the gas phase and aqueous medium. Specifically, in water, compounds 21 ( p -OMe) and 30 ( m,p -diOMe), bearing the strongest donating groups, exhibited shorter bond lengths compared to their results measured in the gas phase.

To describe the lipophilicity of the studied molecular systems, logP values were also calculated 48 , 49 . Compounds 7 ( m,p -diCl), 10 ( o -CF 3 ), 14 ( p - t Bu), and 17 (naphthyl) exhibited the highest logP values, along with their analogous thiophenol pairs: compounds 24 , 27 , 31 , and 34 . However, discrepancies in trends were observed in comparing phenols and thiophenols for compounds 3 ( p -F), 4 ( p -OMe), and 9 ( p -Me), which demonstrated more lipophilic characteristics than their thiophenol counterparts. No meaningful correlation was found between the experimental scavenging results and either bond length or logP values (see SI, Tables S3 – S6 ).

Lastly, Mulliken spin density can help predict stability by showing how the unpaired electron (resulting from the electron donation) is distributed across the molecule. When an antioxidant molecule donates an electron to a free radical, it often becomes a radical itself. The stability of this new radical is crucial for the antioxidant action. A more delocalized spin density (spread over several atoms) generally indicates a more stable radical, enhancing the molecule’s antioxidant efficacy. Thiophenols demonstrate higher spin density values in all cases, the highest being compound 25 ( o -Et), 28 ( p -NO 2 ), 32 (pyridine) and 34 (naphthyl). Comparing theoretical Mulliken spin density values with experimental data on scavenging activity, a significant correlation was found in ABTS assay for thiophenols ( R 2 = 0.84, Fig. 4 g).

The Fukui function has appeared as an effective and more convenient reactivity descriptor in the last few years compare to FMO representations 38 , 52 – 54 . These functions are local descriptors and play a crucial role in assessing the reactivity of a compound by identifying the most nucleophilic or electrophilic site based on electron density (Table S4 , SI). The Fukui function is defined as the derivative of the electron density ( ρ(r) ) concerning the number of electrons ( N ) at a given external potential ( υ ) while maintaining the molecular geometry (Eq. 1 ) 55 . This assumes that chemical reactions are more likely to proceed efficiently when one reactant approaches another at a point where there’s a significant change in the energy associated with the distribution or arrangement of electrons. This is because such fluctuations can lead to more favorable conditions for the reagents to interact, share, or transfer electrons, which is fundamental to most chemical reactions.

where f(r) is the corresponding Fukui function, ρ is the electron density, N is the number of electrons. The term f − corresponds to the system’s response under electrophilic attack (Eq. 3 ), characterized by a decrease in the number of electrons. Conversely, f + denotes the system’s behavior under nucleophilic attack, where it tends to gain electron(s) (Eq. 2 ) 56 .

Condensed Fukui functions and dual descriptors are based on population analysis techniques in order to become quantifiable 57 . Dual descriptors reveal a detailed dual reactivity profile (Eq. 4 ), highlighting areas more susceptible to electrophilic attacks (shown in light blue, Fig. 12 ) as well as to nucleophilic attacks (shown in neon green, Fig. 12 ). The selected examples highlight pairs of analogous compounds, similar to the comparison of HOMOs in Fig. 11 . Compounds 1 and 13 are distinguished as phenols with superior radical scavenging activities, while compounds 32 and 33 are identified as the most efficient thiophenols studied here. Within the first pair of analogues ( 1 and 18 ), an interesting observation emerges, in both instances, the amine functional group is more susceptible to electrophilic attack, outperforming the hydroxyl and thiol groups. Amines are also recognized for their significant antioxidant potential 31 . For compound 13 , oxygen atoms 10 and 8 (hydroxyl) exhibit the most pronounced nucleophilic character, contrasting with compound 30 , where sulfur does not significantly contribute, unlike in compound 32 , where sulfur is the most negative site, surpassing that of nitrogen. Notably, in the analogous compound 15 , nitrogen demonstrates a greater potential for electron donation which can be connected to the lone pair of electrons on the nitrogen. Furthermore, when examining compounds 16 and 33 , the susceptibility to electrophilic attacks appears to be shared between the oxadiazole ring and the hydroxyl/thiol groups, with a bias towards sulfur in the thiophenol and more towards to the nitrogen atoms in the phenol derivative. The weighted condensed Fukui and dual descriptor values are summarized in SI Table S4 .

Based on data presented above several molecular descriptors emerged as effectively and directly influencing the experimental antioxidant effect of phenols and thiophenols (Tables 3 and 4 ). Although at this level of the investigations we are not able to derive one unique function to describe the activity, the contribution of the individual properties provides valuable information on a potential design of more effective antioxidants. The most decisive predictive parameters appear to be the IP and HOMO descriptors, particularly for phenols. Additionally, the dipole moment for phenols and both the BDE and spin density descriptors for thiophenols are critical in the derivatives we studied.

Conclusion

In this study a comparative investigation and analysis of the antioxidant activity profile of 17 phenols and their thiophenol counterparts was described. The experimental antioxidant activities of these compounds were assessed through their scavenging performance in the widely accepted and used DPPH and ABTS assays. Notably, compounds 1 , 4 , 5 , 6 , 13 , and 17 among phenols, and 25 , 26 , 31 , 32 , and 33 among thiophenols, exhibited superior scavenging abilities, in most cases significantly surpassing that of the known reference antioxidant Trolox.

To better understand the mechanisms behind their antioxidant actions extensive DFT calculations were carried out assessing physicochemical properties in both the gas phase and water. In the gas phase, lower BDE values, compared to IP and PA, suggested that the formal-HAT mechanism predominates in the case of these compound groups. In contrast, in water, significant reductions in PA due to solvent effects suggested that the SPLET mechanism is dominant under aqueous conditions. In addition to thermodynamic parameters, other properties such as S–H and O–H bond distances, dipole moments, logP values, and electronic descriptors including HOMO and LUMO orbital energies and gaps were examined. It has been found that several parameters, such as proton affinity, ionization potential, HOMO energy, spin density and even dipole moment had a reasonable correlation with the experimental activities. It is interesting to note that the bond dissociation enthalpy that is often found to play a major role in radical scavenging activity, in the case of these two compound groups did not appear to have a major contribution to the antioxidant activity. Notably, compounds 1 , 13 , 32 , and 33 , representing the most effective phenols and thiophenols in this study, were further analyzed for their condensed Fukui functions, dual descriptor values, and isosurface characteristics. Additionally, to identify the key physical characteristics that contribute to antioxidant efficiency, potential correlations between the calculated properties of phenols, thiophenols, and their experimental radical scavenging activities were explored. The most critical predictive parameters were found to be the ionization potential and highest occupied molecular orbital descriptors, particularly for phenols. For thiophenols, bond dissociation enthalpies and spin density were crucial. Based on the obtained correlations several activity vs. physicochemical parameter functions were created that could be used to predict the antioxidant activity of related compounds. The presented data clearly showed the viability of the analysis, and also confirmed the original hypothesis of using thiophenols as antioxidants. Further research and a larger dataset are needed to develop a more robust and unified quantitative structure–activity relationship (QSAR) model for these compounds. In addition, by using these data as a training set it is planned to apply related compounds (i.e. other substituted phenols and thiophenols), to validate the proposed correlations described in this work that may lead to a better QSAR that have significant predictive values. While in the current study, the most common mechanisms of antioxidant action were investigated, the further analysis of more complex mechanisms, as described in the literature 58 , could expand our understanding of the role of multiple electron and proton transfers to achieve stable radical antioxidants.

Supplementary Information

Supplementary Information.

Acknowledgements

The financial support provided by University of Massachusetts Boston and National Institute of Health (1R15HD110945-01A1) is gratefully acknowledged.

Author contributions

MT and BT conceived the idea, MT secured funding, MT and BT directed the work, M.M. and B.V. carried out experiments, BV, BT, MT wrote the main manuscript text, BV prepared figures. All authors reviewed the manuscript.

Data availability

The materials used in this study were all commercially available products and were purchased from several suppliers as noted in the Supplementary Information. All data generated or analyzed during this study are included in this published article and its Supplementary Information.

Declarations

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-024-83982-4.

References

1. Sies, H. Biochemistry of oxidative stress. Angew. Chem. Int. Ed. 25 , 1058–1071. 10.1002/anie.198610581 (1986).

2. Betteridge, D. J. What is oxidative stress?. Metabolism 49 , 3–8. 10.1016/S0026-0495(00)80077-3 (2000).

3. Sies, H. et al. Oxidative stress. Ann. Rev. Biochem. 86 , 715–748. 10.1146/annurev-biochem-061516-045037 (2017).

4. Sharifi-Rad, M. et al. Lifestyle, oxidative stress, and antioxidants: back and forth in the pathophysiology of chronic diseases. Front. Physiol. 11 , 694. 10.3389/fphys.2020.00694 (2020).

5. Jenner, P. Oxidative stress in Parkinson’s disease. Ann. Neurol. Off. J. Am. Neurol. Assoc. Child Neurol. Soc. 53 , S26–S38. 10.1002/ana.10483 (2003).

6. Markesbery, W. R. Oxidative stress hypothesis in Alzheimer’s disease. Free Radic. Biol. Med. 23 , 134–147. 10.1016/S0891-5849(96)00629-6 (1997).

7. Chang, K.-H. & Chen, C.-M. The role of oxidative stress in Parkinson’s disease. Antioxidants 9 , 597. 10.3390/antiox9070597 (2020).

8. Pintye, D. et al. Nitroxide—HMP—protects human trophoblast HTR-8/SVneo cells from H 2 O 2 -induced oxidative stress by reducing the HIF1A signaling pathway. Antioxidants 12 , 1578. 10.3390/antiox12081578 (2023).

9. Sorriento, D., De Luca, N., Trimarco, B. & Iaccarino, G. The antioxidant therapy: New insights in the treatment of hypertension. Front. Physiol. 9 , 258. 10.3389/fphys.2018.00258 (2018).

10. Charlton, N. C., Mastyugin, M., Török, B. & Török, M. Structural features of small molecule antioxidants and strategic modifications to improve potential bioactivity. Molecules 28 , 1057. 10.3390/molecules28031057 (2023).

11. Platzer, M. et al. Radical scavenging mechanisms of phenolic compounds: A quantitative structure-property relationship (qspr) study. Front. Nutr. 10.3389/fnut.2022.882458 (2022).

12. Dellinger, B. et al. Formation and stabilization of persistent free radicals. Proc. Combust. Inst. 31 , 521–528. 10.1016/j.proci.2006.07.172 (2007).

13. Lu, A. et al. QSAR study of phenolic compounds and their anti-DPPH radical activity by discriminant analysis. Sci. Rep. 12 , 7860. 10.1038/s41598-022-11925-y (2022).

14. Duthie, G. Plant-derived phenolic antioxidants. Curr. Opin. Lipidol. 11 , 43–47. 10.1097/00075197-200011000-00006 (2000).

15. Shi, Y. Support vector regression-based QSAR models for prediction of antioxidant activity of phenolic compounds. Sci. Rep. 11 , 8806. 10.1038/s41598-021-88341-1 (2021).

16. Galano, A. & Alvarez-Idaboy, J. R. Glutathione: Mechanism and kinetics of its non-enzymatic defense action against free radicals. RSC Adv. 1 , 1763–1771. 10.1039/C1RA00474C (2011).

17. Ulrich, K. & Jakob, U. The role of thiols in antioxidant systems. Free Radic. Biol. Med. 140 , 14–27. 10.1016/j.freeradbiomed.2019.05.035 (2019).

18. Silvagno, F., Vernone, A. & Pescarmona, G. P. The role of glutathione in protecting against the severe inflammatory response triggered by covid-19. Antioxidants 9 , 624. 10.3390/antiox9070624 (2020).

19. Carreon-Gonzalez, M. et al. Thiophenols, promising scavengers of peroxyl radicals: mechanisms and kinetics. J. Comput. Chem. 40 , 2103–2110. 10.3390/antiox9070624 (2019).

20. Rimarčík, J. et al. On the enthalpies of homolytic and heterolytic S-H bond cleavage in para and meta substituted thiophenols. Comput. Theor. Chem. 967 , 273–283. 10.1016/j.comptc.2011.04.029 (2011).

21. Klein, E. et al. Study of N–H, O–H, and S–H bond dissociation enthalpies and ionization potentials of substituted anilines, phenols, and thiophenols. J. Mol. Struct. Theochem 758 , 149–159. 10.1016/j.theochem.2005.10.015 (2006).

22. Yan, J. et al. Regenerable thiophenolic radical-trapping antioxidants. Org. Lett. 17 , 6162–6165. 10.1021/acs.orglett.5b03169 (2015).

23. Peerannawar, S. et al. Theoretical and experimental analysis of the antioxidant features of diarylhydrazones. Struct. Chem. 28 , 391–402. 10.1007/s11224-016-0867-x (2017).

24. Shalaby, M. A. et al. Microwave-assisted synthesis, antioxidant activity, docking simulation, and DFT analysis of different heterocyclic compounds. Sci. Rep. 13 , 4999. 10.1038/s41598-023-31995-w (2023).

25. Hesamzadeh, P. et al. De novo antioxidant peptide design via machine learning and DFT studies. Sci. Rep. 14 , 6473. 10.1038/s41598-024-57247-z (2024).

26. Frisch, M. J. et al. Gaussian09 Revision E01 (Gaussian Inc., 2009).

27. Becke, A. D. Density-functional exchange-energy approximation with correct asymptotic behavior. Phys. Rev. A 38 , 3098. 10.1103/PhysRevA.38.3098 (1988).

28. Lee, C. et al. Development of the colle-salvetti correlation-energy formula into a functional of the electron density. Phys. Rev. B 37 , 785. 10.1103/physrevb.37.785 (1988).

29. Rouhani, M. Evaluation of structural properties and antioxidant capacity of Proxison: A DFT investigation. Comput. Theor. Chem. 1195 , 113096. 10.1016/j.comptc.2020.113096 (2021).

30. Hannachi, D., Amrane, N. E. H., Merzoud, L. & Chermette, H. Exploring the antioxidant activity of thiaflavan compounds: A quantum chemical study. New J. Chem. 45 , 13451–13462. 10.1039/D1NJ01996A (2021).

31. Horton, W. et al. Theoretical and experimental analysis of the antioxidant features of substituted phenol and aniline model compounds. Struct. Chem. 30 , 23–35. 10.1007/s11224-018-1183-4 (2019).

32. Tian, Lu. & Chen, F. Multiwfn: A multifunctional wavefunction analyzer. J. Comput. Chem. 33 , 580–592. 10.1002/jcc.22885 (2012).

33. Spiegel, M. et al. A statistically supported antioxidant activity DFT benchmark—the effects of Hartree-Fock exchange and basis set selection on accuracy and resources uptake. Molecules 26 , 5058. 10.3390/molecules26165058 (2021).

34. Mishra, K. et al. Estimation of antiradical properties of antioxidants using DPPH assay: A critical review and results. Food Chem. 130 , 1036–1043. 10.1016/j.foodchem.2011.07.127 (2012).

35. Munteanu, I. G. & Apetrei, C. Analytical methods used in determining antioxidant activity: A review. Int. J. Mol. Sci. 22 , 3380. 10.3390/ijms22073380 (2021).

36. Galano, A. & Raúl Alvarez-Idaboy, J. Computational strategies for predicting free radical scavengers’ protection against oxidative stress: Where are we and what might follow?. Int. J. Quantum Chem. 119 , e25665. 10.1002/qua.25665 (2019).

37. Chen, J. et al. Structure-antioxidant activity relationship of methoxy, phenolic hydroxyl, and carboxylic acid groups of phenolic acids. Sci. Rep. 10 , 2611. 10.1038/s41598-020-59451-z (2020).

38. Spiegel, M. Current trends in computational quantum chemistry studies on antioxidant radical scavenging activity. J. Chem. Inf. Model. 62 , 2639–2658. 10.1021/acs.jcim.2c00104 (2022).

39. Galano, A. et al. Food antioxidants: Chemical insights at the molecular level. Ann. Rev. Food Sci. Technol. 7 , 335–352. 10.1146/annurev-food-041715-033206 (2016).

40. Mittal, A., Vashistha, V. K. & Das, D. K. Free radical scavenging activity of gallic acid toward various reactive oxygen, nitrogen, and sulfur species: A DFT approach. Free Radic. Res. 57 , 81–90. 10.1080/10715762.2023.2197556 (2023).

41. Mittal, A. & Kakkar, R. The antioxidant potential of retrochalcones isolated from liquorice root: A comparative DFT study. Phytochemistry 192 , 112964. 10.1016/j.phytochem.2021.112964 (2021).

42. Mayer, J. M., Hrovat, D. A., Thomas, J. L. & Borden, W. T. Proton-coupled electron transfer versus hydrogen atom transfer in benzyl/toluene, methoxyl/methanol, and phenoxyl/phenol self-exchange reactions. J. Am. Chem. Soc. 124 , 11142–11147. 10.1021/ja012732c (2002).

43. Litwinienko, G. & Ingold, K. U. Solvent effects on the rates and mechanisms of reaction of phenols with free radicals. Acc. Chem. Res. 40 , 222–230. 10.1021/ar0682029 (2007).

44. Galano, A. & Alvarez-Idaboy, J. R. A computational methodology for accurate predictions of rate constants in solution: Application to the assessment of primary antioxidant activity. J. Comput. Chem. 34 , 2430–2445. 10.1002/jcc.23409 (2013).

45. Hansch, C., Leo, A. & Taft, R. A survey of Hammett substituent constants and resonance and field parameters. Chem. Rev. 91 , 165–195. 10.1021/cr00002a004 (1991).

46. Rimarčík, J. et al. Theoretical study of structure and electronic properties of cyano-substituted pyrroles. Chem. Phys. 353 , 177–184. 10.1016/j.chemphys.2008.08.010 (2008).

47. Daina, A., Michielin, O. & Zoete, V. iLOGP: A simple, robust, and efficient description of n-octanol/water partition coefficient for drug design using the GB/SA approach. J. Chem. Inf. Model. 54 , 3284–3301. 10.1021/ci500467k (2014).

48. http://www.swissadme.ch/

49. Daina, A., Michielin, O. & Zoete, V. SwissADME: A free web tool to evaluate pharmacokinetics, drug-likeness and medicinal chemistry friendliness of small molecules. Sci. Rep. 7 , 42717. 10.1038/srep42717 (2017).

50. Pryor, W. A. Why is the hydroxyl radical the only radical that commonly adds to DNA? Hypothesis: It has a rare combination of high electrophilicity, high thermochemical reactivity, and a mode of production that can occur near DNA. Free Radic. Biol. Med. 4 , 219–223. 10.1016/0891-5849(88)90043-3 (1988).

51. León-Carmona, J. R. & Galano, A. Is caffeine a good scavenger of oxygenated free radicals?. J. Phys. Chem. B 115 , 4538–4546. 10.1021/jp201383y (2011).

52. Hossen, J. et al. Theoretical investigations on the antioxidant potential of 2, 4, 5-Trihydroxybutyrophenone in different solvents: a DFT approach. Results Chem. 4 , 100515. 10.1016/j.rechem.2022.100515 (2022).

53. Bardak, F. Experimental and DFT analysis of structural and spectroscopic features of nitroterephthalic acid, and computational insights into its molecular interactions with hER-α via molecular docking. J. Mol. Struct. 1175 , 458–470. 10.1016/j.molstruc.2018.07.110 (2019).

54. Brala, C. J. et al. The average local ionization energy and Fukui function of L-ascorbate, the local reactivity descriptors of antioxidant reactivity. Comput. Theor. Chem. 1049 , 1–6. 10.1007/s00894-014-2476-1 (2014).

55. Parr, R. G. & Yang, W. Density functional approach to the frontier-electron theory of chemical reactivity. J. Am. Chem. Soc. 106 , 4049–4050. 10.1021/ja00326a036 (1984).

56. Parr, R. G. Reviews of Modern Quantum Chemistry: A Celebration of the Contributions of Robert G. Parr Vol. 1, 992–1012 (World Scientific, 2002).

57. Sen, K. D. Reviews of Modern Quantum Chemistry: A Celebration of the Contributions of Robert G. Parr Vol. 1 (World Scientific, 2002).

58. Mittal, A., Vashistha, V. K. & Das, D. K. Recent advances in the antioxidant activity and mechanisms of chalcone derivatives: a computational review. Free Radic. Res. 56 , 378–397. 10.1080/10715762.2022.2120396 (2022).

Associated Data

Supplementary Materials

Supplementary Information.

Data Availability Statement

The materials used in this study were all commercially available products and were purchased from several suppliers as noted in the Supplementary Information. All data generated or analyzed during this study are included in this published article and its Supplementary Information.